The exposure of humans to harmful chemical agents is well documented. Estimates of the cancer incidence due to environmental exposures vary, but it is clear that many industrial chemicals, environmental pollutants as well as dietary components have been implicated as etiological agents of cancer. Currently, however, human risk assessment is generally based on results determined from animal model systems, supplemented by structural modeling and where possible, epidemiological data. Thus, the more informative animal model systems become, the more valuable they become in terms of predicting the risks of human exposure. While a number of short term assays are available to assess the mutagenic potential of these etiological agents in vitro, the assessment of their carcinogenic potential in vivo has required whole animal studies such as the two year bioassay performed by the National Toxicology Program. This assay, however, has been problematic due to the time required for completion, high costs and difficulties in extrapolating the data to risk assessments in humans. This is highlighted by the fact that mice make poor predictors of carcinogenicity in rats. Similarly, rats make poor predictors of carcinogenicity in mice. The extrapolation to human risk is made even more complex as many of these chemicals produce tumors in one tissue in one species and a different tissue in the second species. In many cases the sex of the animal also influences the tissue specificity. This proposal takes advantage of recently developed transgenic mice and rats, containing the lacI transgene for the convenient study of mutagenesis in vivo. These assays are considerably less expensive and much faster than the conventional NTP bioassays. Moreover, as mutational specificity data can also be obtained, mechanistic studies are also possible. Knowledge of the mechanisms by which an agent produces mutation in vivo will be invaluable in determining the potential risk to humans. We propose to use these transgenic rodents to examine the mutational potential and mutational specificity of two selected genotoxic carcinogens, which demonstrate species, gender, and/or tissue specificity. In addition, we will also examine the mutagenic potential of four non-genotoxic carcinogens which are of interest to human health, in one or both species of rodents to determine the robustness of this in vivo assay as a predictor of carcinogenesis. Knowledge of the mutational potency, mutational specificity in target and non-target tissues of genotoxic and non-genotoxic carcinogens in these two species will provide new information about the differences in the mechanism of tumorigenesis in these species and hence, new insights into the extrapolation from mouse, and rat, to man.